Mihael Mohorčič

Carolina Fortuna, Din Mušić, Gregor Cerar et al.

Artificial Intelligence (AI) technologies are moving from customized deployments in specific domains towards generic solutions horizontally permeating vertical domains and industries. For instance, decisions on when to perform maintenance of roads or bridges or how to optimize public lighting in view of costs and safety in smart cities are increasingly informed by AI models. While various commercial solutions offer user friendly and easy to use AI as a Service (AIaaS), functionality-wise enabling the democratization of such ecosystems, open-source equivalent ecosystems are lagging behind. In this chapter, we discuss AIaaS functionality and corresponding technology stack and analyze possible realizations using open source user friendly technologies that are suitable for on-premise set-ups of small and medium sized users allowing full control over the data and technological platform without any third-party dependence or vendor lock-in.

Shih-Kai Chou, Jernej Hribar, Vid Hanžel et al.

Artificial Intelligence (AI) is being incorporated in several optimization, scheduling, orchestration as well as in native communication network functions. This paradigm shift results in increased energy consumption, however, quantifying the end-to-end energy consumption of adding intelligence to communication systems remains an open challenge since conventional energy consumption metrics focus on either communication, computation infrastructure, or model development. To address this, we propose a new metric, the Energy Cost of AI Lifecycle (eCAL) of an AI model in a system. eCAL captures the energy consumption throughout the development, deployment and utilization of an AI-model providing intelligence in a communication network by (i) analyzing the complexity of data collection and manipulation in individual components and (ii) deriving overall and per-bit energy consumption. We show that as a trained AI model is used more frequently for inference, its energy cost per inference decreases, since the fixed training energy is amortized over a growing number of inferences. For a simple case study we show that eCAL for 100 inferences is 2.73 times higher than for 1000 inferences. Additionally, we have developed a modular and extendable open-source simulation tool to enable researchers, practitioners, and engineers to calculate the end-to-end energy cost with various configurations and across various systems, ensuring adaptability to diverse use cases.

Ljupcho Milosheski, Gregor Cerar, Blaž Bertalanič et al.

In recent years, much work has been done on processing of wireless spectrum data involving machine learning techniques in domain-related problems for cognitive radio networks, such as anomaly detection, modulation classification, technology classification and device fingerprinting. Most of the solutions are based on labeled data, created in a controlled manner and processed with supervised learning approaches. However, spectrum data measured in real-world environment is highly nondeterministic, making its labeling a laborious and expensive process, requiring domain expertise, thus being one of the main drawbacks of using supervised learning approaches in this domain. In this paper, we investigate the use of self-supervised learning (SSL) for exploring spectrum activities in a real-world unlabeled data. In particular, we compare the performance of two SSL models, one based on a reference DeepCluster architecture and one adapted for spectrum activity identification and clustering, and a baseline model based on K-means clustering algorithm. We show that SSL models achieve superior performance regarding the quality of extracted features and clustering performance. With SSL models we achieve reduction of the feature vectors size by two orders of magnitude, while improving the performance by a factor of 2 to 2.5 across the evaluation metrics, supported by visual assessment. Additionally we show that adaptation of the reference SSL architecture to the domain data provides reduction of model complexity by one order of magnitude, while preserving or even improving the clustering performance.

Anže Pirnat, Blaž Bertalanič, Gregor Cerar et al.

Non-intrusive load monitoring (NILM) is the process of obtaining appliance-level data from a single metering point, measuring total electricity consumption of a household or a business. Appliance-level data can be directly used for demand response applications and energy management systems as well as for awareness raising and motivation for improvements in energy efficiency. Recently, classical machine learning and deep learning (DL) techniques became very popular and proved as highly effective for NILM classification, but with the growing complexity these methods are faced with significant computational and energy demands during both their training and operation. In this paper, we introduce a novel DL model aimed at enhanced multi-label classification of NILM with improved computation and energy efficiency. We also propose an evaluation methodology for comparison of different models using data synthesized from the measurement datasets so as to better represent real-world scenarios. Compared to the state-of-the-art, the proposed model has its energy consumption reduced by more than 23% while providing on average approximately 8 percentage points in performance improvement when evaluating on data derived from REFIT and UK-DALE datasets. We also show a 12 percentage point performance advantage of the proposed DL based model over a random forest model and observe performance degradation with the increase of the number of devices in the household, namely with each additional 5 devices, the average performance degrades by approximately 7 percentage points.

Ljupcho Milosheski, Gregor Cerar, Blaž Bertalanič et al.

In recent years, the traditional feature engineering process for training machine learning models is being automated by the feature extraction layers integrated in deep learning architectures. In wireless networks, many studies were conducted in automatic learning of feature representations for domain-related challenges. However, most of the existing works assume some supervision along the learning process by using labels to optimize the model. In this paper, we investigate an approach to learning feature representations for wireless transmission clustering in a completely unsupervised manner, i.e. requiring no labels in the process. We propose a model based on convolutional neural networks that automatically learns a reduced dimensionality representation of the input data with 99.3% less components compared to a baseline principal component analysis (PCA). We show that the automatic representation learning is able to extract fine-grained clusters containing the shapes of the wireless transmission bursts, while the baseline enables only general separability of the data based on the background noise.

Shih-Kai Chou, Maice Costa, Mihael Mohorčič et al.

In future large-scale deployments of 6G and beyond networks, collecting timely information, as measured by the Age of Information (AoI) metric, is becoming increasingly important. At the same time, the environmental impact, often characterized by the resulting Carbon Footprint (CF), depends on both the amount of consumed energy and the Carbon Intensity (CI), i.e., the amount of CO$_2$-equivalent emissions produced per unit of consumed energy. Since CI varies over time, minimizing energy is not equivalent to minimizing CF, as a status update with the same energy demand may result in a different carbon cost depending on when it is transmitted. This makes timely status updating a nontrivial scheduling problem. To address this challenge, we formulate carbon-aware status updating as a constrained Markov Decision Process (MDP) that minimizes AoI subject to CF budget, transmission duty-cycle, and channel-capacity constraints. We then propose Sustainable Age-of-Information-Based Timely Status Updating for Hardware-constrained Edge networks (SAOITHE), a Whittle-index-based scheduling solution that enables scalable real-time scheduling. Using real-world CI traces across low-, medium-, and high-CI regions, the results show that SAOITHE remains within the allocated CF budget while achieving lower AoI than baseline policies. Moreover, the gains are around 25% and 20% in low- and medium-CI regions, respectively, and up to 75% in high-CI settings, while preserving scalability.

Gregor Cerar, Halil Yetgin, Mihael Mohorčič et al.

Since the emergence of wireless communication networks, a plethora of research papers focus their attention on the quality aspects of wireless links. The analysis of the rich body of existing literature on link quality estimation using models developed from data traces indicates that the techniques used for modeling link quality estimation are becoming increasingly sophisticated. A number of recent estimators leverage machine learning (ML) techniques that require a sophisticated design and development process, each of which has a great potential to significantly affect the overall model performance. In this paper, we provide a comprehensive survey on link quality estimators developed from empirical data and then focus on the subset that use ML algorithms. We analyze ML-based link quality estimation (LQE) models from two perspectives using performance data. Firstly, we focus on how they address quality requirements that are important from the perspective of the applications they serve. Secondly, we analyze how they approach the standard design steps commonly used in the ML community. Having analyzed the scientific body of the survey, we review existing open source datasets suitable for LQE research. Finally, we round up our survey with the lessons learned and design guidelines for ML-based LQE development and dataset collection.

Irina Barašin, Blaž Bertalanič, Mihael Mohorčič et al.

Time series classification is a relevant step supporting decision-making processes in various domains, and deep neural models have shown promising performance in this respect. Despite significant advancements in deep learning, the theoretical understanding of how and why complex architectures function remains limited, prompting the need for more interpretable models. Recently, the Kolmogorov-Arnold Networks (KANs) have been proposed as a more interpretable alternative to deep learning. While KAN-related research is significantly rising, to date, the study of KAN architectures for time series classification has been limited. In this paper, we aim to conduct a comprehensive and robust exploration of the KAN architecture for time series classification utilising 117 datasets from UCR benchmark archive, from multiple different domains. More specifically, we investigate a) the transferability of reference architectures designed for regression to classification tasks, b) identifying the hyperparameter and implementation configurations for an architecture that best generalizes across 117 datasets, c) the associated complexity trade-offs and d) evaluate KANs interpretability. Our results demonstrate that (1) the Efficient KAN outperforms MLPs in both performance and training times, showcasing its suitability for classification tasks. (2) Efficient KAN exhibits greater stability than the original KAN across grid sizes, depths, and layer configurations, especially when lower learning rates are employed. (3) KAN achieves competitive accuracy compared to state-of-the-art models such as HIVE-COTE2 and InceptionTime, while maintaining smaller architectures and faster training times, highlighting its favorable balance of performance and transparency. (4) The interpretability of the KAN model, as confirmed by SHAP analysis, reinforces its capacity for transparent decision-making.

Ahmad Anaqreh, Shih-Kai Chou, Mihael Mohorčič et al.

Modeling propagation is the cornerstone for designing and optimizing next-generation wireless systems, with a particular emphasis on 5G and beyond era. Traditional modeling methods have long relied on statistic-based techniques to characterize propagation behavior across different environments. With the expansion of wireless communication systems, there is a growing demand for methods that guarantee the accuracy and interpretability of modeling. Artificial intelligence (AI)-based techniques, in particular, are increasingly being adopted to overcome this challenge, although the interpretability is not assured with most of these methods. Inspired by recent advancements in AI, this paper proposes a novel approach that accelerates the discovery of path loss models while maintaining interpretability. The proposed method automates the formulation, evaluation, and refinement of the model, facilitating the discovery of the model. We examine two techniques: one based on Deep Symbolic Regression, offering full interpretability, and the second based on Kolmogorov-Arnold Networks, providing two levels of interpretability. Both approaches are evaluated on two synthetic and two real-world datasets. Our results show that Kolmogorov-Arnold Networks achieve the coefficient of determination value R^2 close to 1 with minimal prediction error, while Deep Symbolic Regression generates compact models with moderate accuracy. Moreover, on the selected examples, we demonstrate that automated methods outperform traditional methods, achieving up to 75% reduction in prediction errors, offering accurate and explainable solutions with potential to increase the efficiency of discovering next-generation path loss models.

Ivana Kesić, Carolina Fortuna, Mihael Mohorčič et al.

Time series segmentation (TSS) is one of the time series (TS) analysis techniques, that has received considerably less attention compared to other TS related tasks. In recent years, deep learning architectures have been introduced for TSS, however their reliance on sliding windows limits segmentation granularity due to fixed window sizes and strides. To overcome these challenges, we propose a new more granular TSS approach that utilizes the Weighted Dual Perspective Visbility Graph (WDPVG) TS into a graph and combines it with a Graph Attention Network (GAT). By transforming TS into graphs, we are able to capture different structural aspects of the data that would otherwise remain hidden. By utilizing the representation learning capabilities of Graph Neural Networks, our method is able to effectively identify meaningful segments within the TS. To better understand the potential of our approach, we also experimented with different TS-to-graph transformations and compared their performance. Our contributions include: a) formulating the TSS as a node classification problem on graphs; b) conducting an extensive analysis of various TS- to-graph transformations applied to TSS using benchmark datasets from the TSSB repository; c) providing the first detailed study on utilizing GNNs for analyzing graph representations of TS in the context of TSS; d) demonstrating the effectiveness of our method, which achieves an average F1 score of 0.97 across 59 diverse TSS benchmark datasets; e) outperforming the seq2point baseline method by 0.05 in terms of F1 score; and f) reducing the required training data compared to the baseline methods.

Ljupcho Milosheski, Gregor Cerar, Blaž Bertalanič et al.

The so-called black-box deep learning (DL) models are increasingly used in classification tasks across many scientific disciplines, including wireless communications domain. In this trend, supervised DL models appear as most commonly proposed solutions to domain-related classification problems. Although they are proven to have unmatched performance, the necessity for large labeled training data and their intractable reasoning, as two major drawbacks, are constraining their usage. The self-supervised architectures emerged as a promising solution that reduces the size of the needed labeled data, but the explainability problem remains. In this paper, we propose a methodology for explaining deep clustering, self-supervised learning architectures comprised of a representation learning part based on a Convolutional Neural Network (CNN) and a clustering part. For the state of the art representation learning part, our methodology employs Guided Backpropagation to interpret the regions of interest of the input data. For the clustering part, the methodology relies on Shallow Trees to explain the clustering result using optimized depth decision tree. Finally, a data-specific visualizations part enables connection for each of the clusters to the input data trough the relevant features. We explain on a use case of wireless spectrum activity clustering how the CNN-based, deep clustering architecture reasons.

Anže Pirnat, Blaž Bertalanič, Gregor Cerar et al.

Location based services, already popular with end users, are now inevitably becoming part of new wireless infrastructures and emerging business processes. The increasingly popular Deep Learning (DL) artificial intelligence methods perform very well in wireless fingerprinting localization based on extensive indoor radio measurement data. However, with the increasing complexity these methods become computationally very intensive and energy hungry, both for their training and subsequent operation. Considering only mobile users, estimated to exceed 7.4billion by the end of 2025, and assuming that the networks serving these users will need to perform only one localization per user per hour on average, the machine learning models used for the calculation would need to perform 65*10^12 predictions per year. Add to this equation tens of billions of other connected devices and applications that rely heavily on more frequent location updates, and it becomes apparent that localization will contribute significantly to carbon emissions unless more energy-efficient models are developed and used. This motivated our work on a new DL-based architecture for indoor localization that is more energy efficient compared to related state-of-the-art approaches while showing only marginal performance degradation. A detailed performance evaluation shows that the proposed model producesonly 58 % of the carbon footprint while maintaining 98.7 % of the overall performance compared to state of the art model external to our group. Additionally, we elaborate on a methodology to calculate the complexity of the DL model and thus the CO2 footprint during its training and operation.

Gregor Cerar, Halil Yetgin, Mihael Mohorčič et al.

Machine learning (ML) has been used to develop increasingly accurate link quality estimators for wireless networks. However, more in-depth questions regarding the most suitable class of models, most suitable metrics and model performance on imbalanced datasets remain open. In this paper, we propose a new tree-based link quality classifier that meets high performance and fairly classifies the minority class and, at the same time, incurs low training cost. We compare the tree-based model, to a multilayer perceptron (MLP) non-linear model and two linear models, namely logistic regression (LR) and SVM, on a selected imbalanced dataset and evaluate their results using five different performance metrics. Our study shows that 1) non-linear models perform slightly better than linear models in general, 2) the proposed non-linear tree-based model yields the best performance trade-off considering F1, training time and fairness, 3) single metric aggregated evaluations based only on accuracy can hide poor, unfair performance especially on minority classes, and 4) it is possible to improve the performance on minority classes, by over 40% through feature selection and by over 20% through resampling, therefore leading to fairer classification results.

Gregor Cerar, Aleš Švigelj, Mihael Mohorčič et al.

Multiple-input multiple-output (MIMO) is an enabling technology to meet the growing demand for faster and more reliable communications in wireless networks with a large number of terminals, but it can also be applied for position estimation of a terminal exploiting multipath propagation from multiple antennas. In this paper, we investigate new convolutional neural network (CNN) structures for exploiting MIMO-based channel state information (CSI) to improve indoor positioning. We evaluate and compare the performance of three variants of the proposed CNN structure to five NN structures proposed in the scientific literature using the same sets of training-evaluation data. The results demonstrate that the proposed residual convolutional NN structure improves the accuracy of position estimation and keeps the total number of weights lower than the published NN structures. The proposed CNN structure yields from 2cm to 10cm better position accuracy than known NN structures used as a reference.